|Publication number||US4644479 A|
|Application number||US 06/636,195|
|Publication date||Feb 17, 1987|
|Filing date||Jul 31, 1984|
|Priority date||Jul 31, 1984|
|Also published as||CA1237794A, CA1237794A1, DE3582625D1, EP0170516A2, EP0170516A3, EP0170516B1|
|Publication number||06636195, 636195, US 4644479 A, US 4644479A, US-A-4644479, US4644479 A, US4644479A|
|Inventors||Christian T. Kemper, James C. Bellows, Pamela J. Kleinosky|
|Original Assignee||Westinghouse Electric Corp.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (6), Non-Patent Citations (2), Referenced by (191), Classifications (17), Legal Events (8)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Invention
The invention in general relates to diagnostic apparatus, and particularly to a sensor-based system for on-line realtime monitoring.
2. Description of the Prior Art
Complex industrial or other operating systems generally have a plurality of sensors for monitoring various parameters during operation, not only for control purposes but for purposes of monitoring the system to detect actual or impending malfunctions.
Some systems may utilize dozens, if not hundreds, of sensors in the diagnostic process and very often the sensors may fail, degrade, or provide spurious readings not related to the actual parameter being measured.
Use of erroneous sensor data in the diagnostic process can lead to erroneous conclusions about possible malfunctions. In one respect, a malfunction may be indicated where, in fact, no malfunction exists and conversely a malfunction may be occurring or may occur without its detection and without proper notification to the system operator. Such event can represent a tremendous economic loss as well as a potentially dangerous situation.
To obviate these potential problems, some systems utilize redundant sensors such that if one fails another may take its place. For systems utilizing hundreds of sensors, such solution may be unfeasible, from an economic standpoint. In other types of systems, sensor readings are preprocessed prior to the system diagnosis and eliminated from consideration if their readings exceed certain boundaries. With such an arrangement, however, valuable information relative to the sensor degradation history which may be utilized in the diagnostic process, is lost.
The present invention provides for a diagnostic system which can recognize operating problems while they may be little more than a vague trend, and may do so taking into account sensor degradation or failure.
Diagnostic apparatus is provided for monitoring an operating system having a plurality of sensors throughout the system and which generate output signals indicative of certain system parameters. A control establishes a first subsystem for each of a number of selected sensors and includes means to periodically obtain readings of the sensor output signals to provide a plurality of indications characterizing movement, if any, of the sensor signal. Such movement indication may include whether or not a sensor signal has sharply increased or decreased in a first period of time, whether it has slowly increased or decreased in a second and greater period of time or whether or not the signal is steady during the time periods.
The control is additionally operable to establish a second subsystem which is responsive to the sensor readings as well as selected ones of the movement indications to provide validated conclusions relative to the sensor signal with each conclusion having a certain confidence factor in the validity of the conclusion.
The control also establishes a third subsystem which is responsive to the sensor signals as well as the validated conclusions obtained by the second subsystem to provide indications of possible malfunctions in the operating system whereby such malfunctions, with a certain degree of confidence, may be communicated to an operator.
An example of diagnostic apparatus in accordance with the present invention is illustrated with respect to a steam turbine-generator power plant which utilizes a plurality of sensors for obtaining data relative to chemical parameters at the influent and effluent of chemically active components to predict possible malfunctions of these components.
FIG. 1 illustrates a simplified diagram of a steam turbine generator power plant;
FIGS. 2 and 3 illustrate nodal diagrams utilized to explain one type of expert system which may be utilized in the operation of the diagnostic apparatus;
FIGS. 4A and 4B illustrate various functions associated with components of FIGS. 2 and 3;
FIG. 5 is a nodal diagram subsystem illustrating the propagation of belief relative to certain parameters of any sensor utilized in the arrangement of FIG. 1;
FIG. 6 is a nodal diagram subsystem illustrating the propagation of belief of certain characteristics relative to a particular sensor of the arrangement of FIG. 1;
FIG. 7 is a simplified diagram of the type of sensor used in FIG. 6;
FIG. 8 is a nodal diagram subsystem illustrating the propagation of belief of certain characteristics relative to another particular sensor of the arrangement of FIG. 1;
FIG. 9 is a simplified diagram of the type of sensor used in FIG. 8;
FIG. 10 is a nodal diagram subsystem illustrating the propagation of belief relative to a certain malfunction of a component of FIG. 1; and
FIG. 11 illustrates one type of display for presentation of possible malfunctions.
Although the present invention is applicable to a variety of operating systems, it will be described in this preferred embodiment with respect to a steam turbine generator power plant such as illustrated in FIG. 1.
The plant includes a steam turbine arrangement 10 having a plurality of turbines in the form of a high-pressure turbine 12, intermediate pressure turbine 14 and low pressure turbine 16, all of which are coupled to a common shaft 18 to drive an electrical generator 20 which supplies power to a load 22 when on-line with main circuit breakers 23 closed.
A steam supply such as a fossil-fired once-through boiler system 24 includes, by way of example, and input economizer section 26, a superheater 27 and a reheater 28. Boiler steam is provided to the turbine arrangement 10 through input valving 30 and steam exiting the high-pressure turbine 12 is reheated in reheater 28 and provided to intermediate pressure turbine 14 through valving 32. Steam exiting the intermediate pressure turbine 14 is provided by way of cross-over piping 34 to the low-pressure turbine 16 from which the steam is exhausted into a conventional condenser 36 in heat exchange relationship with externally supplied cooling water.
Water in the condenser is recirculated back to the boiler after chemical treatment to maintain high purity. The chemical treatment may include a plurality of condensate polishers 40 which basically are ion exchange units designed to remove certain impurities. After the chemical treatment, the water is heated by a series of heaters 42 including a deaerating heater which removes dissolved gases, and, after addition of certain chemicals, is returned to the input economizer 26 of the boiler system 24.
The power plant is provided with a plurality of sensors including sensors to monitor chemical parameters of the steam in the turbine system as well as condensate in the path between the condenser and boiler system. Accordingly, sensor array 50 is provided and includes a plurality of sensors S1, S2 . . . Sn which receive sample steam from the steam path and reduced to suitable pressure and temperature by conditioner 52. Similarly, a plurality of other sensor arrays 54 to 56 are respectively provided at the output of condenser 36, after the polishers 40 and prior to economizer 26. By way of example, the sensors of each array may include those for measuring acid cation exchanged conductivity, sodium concentration, dissolved oxygen, specific conductivity, pH and chloride concentration.
The sensor arrays are positioned at the influent and effluent of chemically active components of the steam turbine system. For example, sensor array 50 in the steam path measures the influent to condenser 36 while sensor array 54 measures the effluent. Sensor array 54 in turn provides data relative to the influent to condensate polishers 40 while sensor array 55 measures the effluent therefrom. Data relative to the influent and effluent of chemical feeds, and heaters 42 are provided by respective sensor arrays 55 and 56 while sensor arrays 56 and 50 provide information relative to the influent and effluent of boiler system 24. Thus, the sensors can provide not only indications of corrosive and other impurities in the system but are strategically located to provide indications of plant component malfunctions as well.
The malfunction assessments are provided by a digital computer 58 controlling the diagnostic process so as to provide possible malfunction indications which may be communicated to an operator via an output 60 such as an alarm system, CRT or other display, message system or combinations thereof.
In a preferred embodiment, the computer controls the diagnostic process by implementation of an expert system computer program that uses knowledge representations and inference procedures to reach conclusions normally determined by a human expert. A common form of knowledge representation is in the form of IF . . . THEN rules and one such system which may be utilized in the practice of the present invention is PDS (Process Diagnosis System) described in the Proceedings of the Eight International Joint Conference on Artificial Intelligence, Aug. 8-12 1983, pages 158-163. Basically, in that system (as well as other expert systems) for each rule there is an antecedent or evidence (the IF portion) as well as a consequent or hypothesis (the THEN portion) which can become evidence for other rules. As depicted in FIG. 2, evidence 64 is linked to the consequent hypothesis 65 by means of rule 66, with the evidence and hypothesis constituting nodes of the system. Numeral 68 represents a supporting rule of node 64, that is, a rule for which the node 64 is a hypothesis. Rule 66 is a supported rule of node 64, that is, a rule for which the node 64 is evidence. Likewise, rule 66 is a supporting rule for node 65. In the system, by way of example, nodes can take the form of evidence, hypothesis, malfunctions, sensors and storage-nodes which are nodes capable of storing values input from other nodes and performing some predetermined mathematical operation on the values.
Associated with each node is a measure of belief, MB, that the node (hypothesis) is true, as well as a measure of disbelief, MD, which is a measure of belief that the hypothesis is not true. Both factors range on a scale from 0 to 1 and the difference between them, MB-MD, yields a certainty or confidence factor CF which ranges from -1 to +1, where positive numbers represent confidence that the hypothesis is true and negative numbers represent belief that the hypothesis is not true; numbers in the vicinity of 0 represent uncertainty.
An expert (or experts) in the field to which the diagnosis pertains establishes the various rules and relationships which are stored in the computer's memory and utilized in the diagnostic process. The expert's belief in the sufficiency of the rule is also utilized. This belief, which represents the experts opinion as to how the presence of evidence proves the hypothesis, is given a numerical representation designated as a sufficiency factor, SF, which ranges from -1 to +1, where positive values of SF denote that the presence of the evidence suggests that the hypothesis is true and negative values denote that the presence of the evidence suggests that the hypothesis is not true.
PDS additionally utilizes the expert's belief in the necessity of the rule, which illustrates to what degree the presence of the evidence is necessary for the hypothesis to be true. This necessity belief is given a numeral representation designated as a necessity factor NF which ranges from -1 to +1 where positive values of NF denote that the absence of evidence suggests that the hypothesis is not true and negative values denote that the absense of the evidence suggests that the hypothesis is true.
FIG. 3 illustrates another common arrangement wherein a plurality of rules 68 to 70 connect evidence nodes 72 to 75 to a hypothesis node 76. Element 78 represents the combining of evidence in (a) a disjunctive manner, that is, if evidence 74 OR 75 is present, or (b) in a conjunctive manner, that is, if evidence 74 AND 75 are present.
Belief leading to a consequent possible malfunction in the system being diagnosed is propagated from evidence to hypothesis in repetitive cycles, at the beginning of which the CF, MB and MD values of each node are reset to zero (except for a sensor node where the MB and accordingly the CF is assumed to be +1).
If the CF of the evidence is positive, then the rule's sufficiency is utilized to propagate belief, whereas if the CF of the evidence is negative, the rule's necessity is utilized; if CF is zero, nothing is done.
Basically, if the evidence CF is positive and the SF is positive, then the MB of the hypothesis is increased; if the SF is negative, then the MD of the hypothesis is increased.
Conversely, if the evidence CF is negative, and the NF positive, then the MD of the hypothesis is increased, and if the NF is negative, the MB of the hypothesis is increased. By way of example, for the single rule case of FIG. 2, if MB and MD are the belief and disbelief in the rule's hypothesis, CF the confidence in the rule's evidence, and SF and NF are the rule's sufficiency and necessity, then:
if CF>0 and SF>0:MB=CF×SF (1)
if CF>0 and SF<0:MD=CF×(-SF) (2)
if CF<0 and NF>0:MD=(-CF)×NF (3)
if CF<0 and NF<0:MB=CF×NF (4)
For the multiple rule case of FIG. 3, final values are obtained by examining each rule in sequence and performing the calculations for each rule in accordance with the following, where MBold and MDold are the belief and disbelief in the rule's hypothesis before each calculation, CF the confidence in the rule's evidence, SF and NF are the rule's sufficiency and necessity and MBnew and MDnew are the belief and disbelief in the rule's hypothesis after each calculation:
if CF>0 and SF>0:MBnew =MBold +(1=MBold)×CF×SF (5)
if CF>0 and SF<0:MDnew =MDold +(1-MDold)×CF×(-SF) (6)
if CF<0 and NF>0:MDnew =MDold +(1-MDold)×(-CF)×NF (7)
if CF<0 and NF<0: MBnew =MBold +(1-MBold)×CF×NF (8)
For a disjunctive logical node (OR function) then the highest confidence factor of all of the pieces of evidence may be utilized, whereas if the logical node is conjunctive (AND function) the minimum of all of the confidence factors may be utilized. Alternatively, a weighted average may be utilized.
Thus, by utilizing the appropriate previous equations, a measure of belief and/or disbelief is calculated for a hypothesis and from these values a confidence factor in the hypothesis is calculated from the relationship CF=MB -MD.
A rule's sufficiency (SF) or necessity (NF) may in many instances by expressed as a constant. In other instances, the sufficiency and/or necessity may be expressed as some other function which will generate a sufficiency or necessity factor of a fixed number by evaluating the function for a particular variable. A common function which may be utilized is a piece-wise linear function, two examples of which are illustrated in FIGS. 4A and 4B. The Y-axis in these figures represent the SF (or NF) ranging from -1 to +1 on the vertical scale. The X-axis horizontal scale represents the value of some variable such as a sensor reading or the result of some mathematical operation, by way of example. In FIG. 4A, if the variable has a value between 0 and a, or is greater than f, it will generate an SF of -1 whereas if the value is between c and d, it will generate an SF of +1. Values between a and c or d and f will generate corresponding SFs between -1 and +1. FIG. 4B represents a piece-wise linear function wherein any variable value greater than b will generate an SF of +1, any variable value less than -b will generate an SF of -1 and values between -b and +b will generate a corresponding SF between -1 and +1.
Another type of useful rule is a reading-transform rule which, when carried out, applies a transform function to the value found in the rule's evidence node. If the evidence node is a sensor, the value is a sensor reading, with appropriate conversion, scaling, etc. performed by the transform, if needed.
In FIG. 1, the various types of sensors utilized to gather data are themselves subject to erroneous operation and therefore may lead to an erroneous diagnosis of plant components. In accordance with the present invention, the computer controlling the diagnosis operation establishes a first subsystem which is generic to all sensors and which may be instantiated, or applied, with respect to selected sensors throughout the system. A typical generic subsystem utilizing the expert system nodal and rule concept is illustrated in FIG. 5. The arrangement is responsive to the sensor output signals to provide an indication of sensor signal movement. In FIG. 5 and subsequent Figs. an item designated as a sensor is a node into which actual sensor readings are placed.
Sensor 100 periodically provides sensor readings which are linked by a reading transform rule 102 to node 104 which is operable to obtain an average of the last five sensor readings so as to provide some stability and filter out any sensor noise. At the other extreme, reading transform rule 106 links the sensor readings to node 108 which is operable to obtain the value of the current sensor reading minus the previous sensor reading so as to be sensitive to sensor variation and provide an indication of just how much noise is present in the system.
Baseline function 110, linked to the sensor by reading transform rule 112 is operable to provide a relatively longer term indication of sensor performance. Let it be assumed, merely by way of example, that sensor readings are periodically provided once a minute. Baseline function 110 may be operable to accumulate readings over a period of time such as 30 minutes and provide the average of, for example, the first 10 readings of the 30 minute period.
Accordingly, function 108 provides indications of changes from minute to minute, function 104 provides a 5 minute average and baseline function provides an average over an interval in past time.
The last 5 and baseline functions are utilized to obtain various indications of short term or long term sensor performance. Node 114 determines the change, if any, between the current sensor reading provided by rule 116 and the average of the last 5 provided by rule 118. The amount of change relative to the average is determined at node 120 which functions to divide the value of change by the average of the last five readings. To accomplish this, the relative change node 120 is linked by rule 122 to node 114 and by rule 124 to node 104.
Trending information, or slope, is obtained by node 126 which is linked to the last five average node 104 by rule 128 and to baseline node 110 by rule 130. Functionally, the slope is obtained by subtracting the baseline value from the last five average and dividing by the time period between the centers of the data group. That is, the time period between the center of the ten readings taken 30 to 20 minutes ago and the center of the last five readings is 22.5 time units or 221/2 minutes in the present example.
In addition to the slope, the relative slope is obtained at node 132 by dividing the slope, linked by rule 134, by the baseline, linked by rule 136.
If the relative change, as determined at node 120, is positive, its belief is propagated to node 138 indicative of a positive relative change, with the propagation being via rule 140. Similarly, belief in a negative relative change is propagated to node 142 via rule 144.
In an analogous manner, belief in a positive relative slope at node 132 is propagated to node 146 via rule 148 and belief in a negative relative slope is propagated to nodeU 150 via rule 152.
Accordingly, nodes 138 and 142 set forth information with a certain degree of confidence with respect to relatively short term sensor changes. Nodes 146 and 150 set forth information with a certain degree of confidence with respect to relatively longer term changes in sensor readings. Confidence in these beliefs, however, will be lowered if the sensor readings are close to the detection limit of the sensor itself in which case, certain values may become unstable militating against drawing large conclusions based on these values.
In order to reduce the confidence factors associated with nodes 138, 142, 146 and 150, the subsystem includes means for adding disbelief to these nodes via rules 154 to 157 should be sensor be operating near its detection limit. Node 160 determines the ratio of the average sensor reading compared to the detection limit, with the average being propagated by rule 162 from node 104 and the detection limit being provided by a detection limit node 164 via rule 166. If the ratio calculated at node 160 is less than one, for example, belief is propagated to node 168 via rule 170 that the apparatus is in fact near the detection limit so as to modify belief in the short term and long term changes.
If, in fact, operation is not near detection limits and a positive relative change has occurred, then this is indicative of a short term rise in the sensor output as determined at node 172 linked to node 138 by rule 174. A positive change propagated from node 114 by rule 175 would also indicate this short term rise. If a negative relative change has occurred, then this is indicative of a short term drop as indicated at node 176 connected to node 142 by rule 178. A negative change propagated from node 114 by rule 179 would also indicate this short term drop.
If, over the long term, the relative slope is positive as indicated at node 146, then this condition is indicative of a slowly increasing sensor output as depicted at node 180 connected to node 146 by rule 182. A positive slope propagated from node 126 by rule 183 would also indicate this long term increase. Conversely, a negative relative slope gives rise to a decreasing sensor output indication at node 184 conencted to node 150 by rule 186. A negative slope propagated from node 126 by rule 187 would also indicate this long term decrease.
The short term nodes 172 and 176 and the long term nodes 180 and 184 are evidence for whether or not this sensor output is steady as defined by node 188 and connected to these previous nodes by respective rules 189 to 192. All of these latter rules would have negative sufficiency functions creating disbelief in the steady hypothesis, since, if the sensor output is rising or dropping or increasing or decreasing, it is not steady.
A positive belief in a steady situation may be propagated from node 194 connected to steady node 188 by rule 196 and indicative of a small variation. That is, if the sensor output signal does not change from reading to reading, or changes by a very small predetermined amount, then this situation is indicative of a steady condition. In order to determine this variation, the change from reading to reading as developed at node 108 is propagated via rule 200, which takes the absolute value of the change, and node 198 derives the value for average change. A piece-wise linear function associated with rule 202 would then propagate belief or disbelief to node 194 depending upon the value of the average change.
Many sensors can operate over a wide dynamic range and accordingly include scale changing facilities. If identical sensors throughout the system have been set to the same scale and one sensor subsequently and inadvertently is switched to a different scale, then its output contribution would be misleading. Accordingly, the subsystem of FIG. 5 includes means for determining whether or not a scale change has been made. This is accomplished with the provision of node 204 which determines the ratio of the average of the last five readings from node 104 via rule 206 to the baseline value from node 110 via rule 208. Depending upon the value derived at node 204, a piece-wise linear function associated with rule 210 would propagate the belief that there has or has not been a scale change. By way of example, suppose that the sensor instrument includes means for changing the scale by factors of ten. The piece-wise linear function associated with rule 210 therefore would be +1 for an average to baseline ratio of around 0.1, 1, 10, 100, etc., and -1 for other ratios.
Accordingly, the subsystem of FIG. 5 derives a great deal of information relative to the operation of any sensor to which the subsystem is applied and this information, reproduced at the bottom of FIG. 5, is utilized in subsequent subsystems established by the computer control.
One such subsystem is illustrated in FIG. 6 which pertains, by way of example, to the obtaining of validated readings from a cation conductivity sensor. The subsystem of FIG. 6 is generic to any cation conductivity sensor which may be utilized in the arrangement of FIG. 1 be it for steam measurement or condensate measurement.
Very basically, a cation conductivity sensor, more particularly, an acid cation exchanged conductivity sensor, is utilized to exchange cations in a sample to hydrogen so that anions, commonly associated with corrosion, can be measured. FIG. 7 illustrates a simplified cation conductivity sensor. The sensor includes a cation resin bed 222 which removes the cations present in the sample and replaces them with hydrogen ions. Thus, any sodium chloride present in the sample is converted to hydrochloric acid, sodium sulfate is converted to sulfuric acid, sodium acetate is converted to acetic acid, etc. A conductivity cell 224 then provides an output signal proportional to the conductivity of the cation exchanged sample to provide an indication of anion concentration.
Returning once again to FIG. 6, the subsystem includes several nodes for characterizing the sensor output signal. A determination of whether or not the sensor reading is low is made at node 230 connected to the sensor by rule 232 and the determination of whether or not the sensor reading is high is accomplished at node 234 connected to the sensor by rule 236. Most sensors have provisions for testing the sensor electronics and under such test condition, any output signal provided is in an abnormally high range. Node 238 tests to see whether or not the sensor is in this abnormally high range and is connected to the sensor by rule 240.
If the sensor reading is low, it may be an indication that someone has turned the sensor off (many sensors will still provide a small output when turned off). Belief in the sensor being off may be generated at node 242 linked to node 230 by rule 244. If, however, the sensor reading is not steady, it means that the output readings are changing which strongly suggests that the sensor would not be in an off condition. Node 188 linked to the sensor off node 242 by rule 246 provides this indication of whether or not the sensor reading is steady. This indication is obtained from the steady indication of the subsystem of FIG. 5 as now applied to the specific cation conductivity sensor 220 of FIG. 6. In FIG. 6, as well as in subsequent Figs., nodes utilized from other subsystems are illustrated as dotted rectangles.
A further indication of the sensor being in an off condition is whether or not the sensor reading is equal to the system average as determined at node 248 linked by rule 250 to node 242. In order to make this determination, a node 252 provides an indication of the average readings of all cation conductivity sensors in the system of FIG. 1 and links it by rule 254 to node 256 which is linked to this one particular sensor's output by rule 258 in order to derive an indication of the difference between this sensor's reading and the system average. The result is linked to node 248 by rule 260 which would have a positive sufficiency factor if the results of the subtraction process of node 256 were within a predetermined range of zero.
Another factor which may yield a low sensor reading would be a contaminated or dirty sensor. For example, due to the operating environment, the conductivity electrodes may get contaminated such as by oil which then acts as an insulator resulting in an output reading which is not as high as it should be. This may be determined by the dirty sensor node 262 linked to the low sensor reading node 230 by rule 264. A steady indication linked by rule 266 from node 188 may add to the confirmation of a dirty sensor, however, if the reading is not steady, it does not mean that the sensor isn't dirty. Confidence may be lost in a dirty sensor diagnosis if the sensor reading is equal to the system average with the dirty sensor node 262 being linked to this prior node by rule 268.
Belief that the sensor is in the test range is propagated by rule 270 to node 272 which will confirm with a certain degree of confidence that the sensor is in a test mode taking into account whether or not the reading is steady and whether or not it equals the system average, these latter two factors being propagated by respective rules 274 and 276.
Three sensor conditions, off, dirty, and test, which would lead to decreased confidence in a valid sensor reading have been described. Other factors are also present including the determination of whether or not the sensor is on the correct scale. To make this determination, an OR function 280 is applied to nodes 172 and 176 indicative of a sensor reading rise or drop to determine whether or not there has been a recent sharp sensor reading change as determined at node 282 linked to the OR function by rule 284. With knowledge of a recent sharp sensor reading change propagated by rule 286 and belief as to whether or not a factor of 10 scale change is present propagated from node 212 by rule 288, node 290 may then make the appropriate scale change determination.
If the cation resin of the sensor illustrated in FIG. 7 is used up (exhausted) then ammonia, previously blocked by the resin will cause a relatively high output signal to be provided and which signal will be unrelated to acid cation conductivity measurement. Accordingly, rule 292 propagates the belief in a high sensor reading from node 234 to node 294 for the determination of resin exhaustion. To substantiate that the sensor reading is high, it is compared with the system average. This step was previously done at node 256 with the belief in a higher reading being propagated by rule 296 to node 298 establishing the belief that the sensor reading is in fact higher. This belief, propagated by rule 300 adds to the belief that the sensor resin is probably exhausted.
Two other factors may also lead to this conclusion, one being the fact that the sensor reading is increasing, propagated by node 180 via rule 302. The other input in the determination is via rule 304 from node 306 which establishes that the sensor reading is less than the specific conductivity of the sample. The specific conductivity value is provided by a specific conductivity sensor 310 included as one of the plurality of sensors in the sensor array. This specific conductivity value propagated by rule 312 and the cation conductivity sensor reading propagated by rule 314 are compared at node 316 wherein the specific conductivity value is subtracted from the sensor value with the result being propagated via rule 318 to node 306. If the resin is exhausted, then the value of the cation conductivity sensor output can get no higher than the specific conductivity sensor output. If there is evidence that the cation conductivity sensor value is not less than or equal to the specific conductivity value, then a measure of disbelief will be added to the determination of sensor resin exhaustion at node 294.
By way of example, five possible sensor malfunctions have been presented, namely, a scale change 290, sensor off 242, dirty sensor 262, sensor test 272 and sensor resin exhausted 294. Each malfunction has two or more inputs each lending a certain degree of concern relative to the existence of the malfunction such that the presence of the malfunction can be established with a certain confidence factor ranging from -1 (definitely not present) to +1 (definitely present).
The highest confidence in any of the possible malfunctions may be propagated by OR function 320 and rule 321 to node 322 indicative of a sensor malfunction. Since the sensor conclusions may be utilized in other subsystems to determine plant component malfunctions, the use of results from a failing or failed sensor would impact on accurate plant component malfunction predictions. For example, if there is a high sensor reading as determined at node 234, a logical conclusion is that there is a high anion concentration as specified at node 324 linked to the high sensor reading node 234 by rule 326. From an increasing reading determination made by the subsystem of FIG. 5, results propagated by rule 328 raises the belief in an increasing anion concentration specified by node 330.
Normally, in the absence of any sensor malfunction, confidence in a high anion concentration would be propagated unaltered by rule 332 to a validated high anion concentration node 334. Similarly, 100% of the confidence in an increasing anion concentration of node 330 would be propagated by rule 336 to a validated increasing anion concentration node 338. In the presence of a sensor malfunction, however, the confidence in these results must be reduced. One way of accomplishing this reduction in confidence is by a technique further described and claimed in copending application Ser. No. 061636,196, filed concurrently herewith, assigned to the same assignee as the present invention and hereby incorporated by reference.
Very basically, the change is accomplished by a paralt (parametric alteration) rule which is operable to change the sufficiency function and/or necessity function of another rule. If, in FIG. 6, node 322 indicates a sensor malfunction, paralt rule 350 will modify the sufficiency function of rule 332 to a degree depending upon the confidence in the sensor malfunction such that the confidence in the validated high anion concentration of node 334 is based on the sufficiency function of the modified rule. Similarly, paralt rule 352 will modify the sufficiency function of rule 336 so that confidence in a validated increasing anion concentration of node 338 is based upon the modified values of the rule. The necessity functions are changed in a similar manner.
FIG. 8 illustrates a generic subsystem for validating the results of another type of sensor, namely a sodium sensor 356. The subsystem of FIG. 8 may be instantiated with respect to any sodium sensor or any array throughout the power plant be it connected to a steam line or a liquid line.
A simplified diagram of a typical commercially available sodium sensor is illustrated in FIG. 9. A sodium ladened input sample may be directed to a sensor path 358 or a polisher path 360 by means of a valve 362. When on the sensor path, the sample is directed through tubing 364 passing through an injection port 366. Tubing 364 is immersed in a concentrated ammonia environment 368 so that ammonia diffuses through tubing 364 at a predetermined rate to accurately control the pH value of the sample. The sample is then provided to a set of electrodes 370 including a sodium ion selective electrode and a reference electrode to derive an output signal, the value of which represents parts per billion of sodium concentration.
When valve 362 is operated to connect the input sample to the polisher path 360 for purposes of calibration, polisher 372 removes almost all traces of sodium in the fluid supplied to tubing 364 and the output signal would then reflect a lack of or an extremely low sodium concentration. Calibration is then accomplished by injecting, at port 366, a known concentration of sodium as represented by arrow 374.
Referring once again to FIG. 8, the subsystem obtains validated high and/or increasing sodium concentration indications as a function of the operating condition of the sensor.
High and low sensor reading indications in terms of sodium concentration (parts per billion) are obtained at respective nodes 380 and 382 connected by respective rules 384 and 386 to sensor 356. The low sodium concentration indication is utilized in conjunction with long term trending information to determine the condition of the sodium sensor itself.
In order to obtain this long term trending information, three indications from the first subsystem of FIG. 5, as applied to this particular sodium sensor, are utilized, such indications being a sharp drop indication 176, a steady indication 188 and the last five average reading 105. Time is propagated in the subsystem of FIG. 8 from a time sensor node 390. In the absence of a drop, current time is propagated via rule 392 to be continuously updated by node 394 connected to the drop indication by a paralt rule 396. If a drop does occur, another paralt rule 398 cuts off propagation of time via rule 392 such that node 394 stores the time, T'0, at which the sensor reading drop occurred. Node 400 receives time T0 propagated by rule 402 and the current time propagated by rule 404 and functions to continuously obtain the duration of time since the drop occurred. Paralt rule 396, which fires when there is a sensor reading drop, functions to keep node 394 in an updated state, in the absence of which rule 402 would not fire to propagate time T0 to node 400.
In a similar fashion, if the sensor reading becomes steady, paralt rule 406 cuts off transmission of time via rule 408 to node 410 which, made valid by paralt rule 412, stores the intial time T'0 at which the steady indication occurred. This time, propagated by rule 414 is subtracted from the current time propagated by rule 416, in node 418 to obtain an indication of the duration of the steady reading.
If a recent drop has occurred, there may be some indication that the sodium sensor has been switched to the polisher path. If the reading becomes steady and in the low sensor range, further evidence of sensor polisher path operation is indicated. To make this determination, node 420 receives the duration of steady reading from node 418 via rule 422, the time since the sensor reading drop from node 400 via rule 424, and the low sensor reading from node 382 via rule 426. A further input in the determination is communicated via rule 430 from node 432 which derives an indication of whether or not the particular sensor reading is equal to the system average. This indication is obtained in a manner similar to that of FIG. 6 by providing an average of all the sodium sensor outputs in the system, from node 434 to node 436 via rule 438. The sensor reading itself is communicated via rule 440 to the node which then subtracts the system average from the sensor reading and provides the results to node 432 via rule 442. If, in fact, the sensor reading is equal to the system average, then this would contribute disbelief in the idea that the sensor was on the polisher path.
The arrangement of FIG. 8 also provides relatively long term trending information. The average of the last five readings is communicated via rule 444 to node 446 which stores the values over a 4-hour period, for example, and takes the oldest average from 4 hours ago and subtracts it from the latest average to obtain the change over the 4 hours. The relative 4 hour trend is obtained at node 448 connected to the 4 hour trend node 446 by rule 450 and connected to the average of the last 5 node 105 by rule 452. Information relative to the trend propagated by rule 454 and the relative trend propagated by rule 456 will allow node 458 to determine whether or not the trend is positive. If the drive is positive and steady (i.e. very small slope and few or no irregular variations in amplitude) then this may lead to the conclusion that the tubing in the sodium sensor has ruptured and ammonia is leaking into the sample stream. Node 460 connected to node 418 by rule 462 determines from the information of mode 418 whether or not there has been 4 hours of steady sensor reading. This fact, together with the positive trend are coupled to AND function 464 such that when both are present, rule 466 will propagate the belief to node 468 that there has been a rupture in the tubing. This belief is tempered, however, by disbelief propagated via rule 470 indicative of the sensor reading being equal to the system average.
OR function 472 takes the highest confidence of a possible tubing rupture or sensor being on the polisher path and propagates it via rule 474 to node 476 for determination of whether or not a sensor malfunction has actually occurred. In the absence of any malfunction, rule 480 propagates the belief in a high sodium concentration to node 482 which establishes a validated belief. Similarly, an increasing sensor reading from node 180 is propagated by rule 484 to node 486 providing an indication of validated increasing sodium concentration.
If not 476 indicates that there has been a sensor malfunction, a paralt rule 488 will modify the sufficiency and necessity functions of rule 480 to change the belief in a validated high sodium concentration. Similarly, with the presence of a malfunction, paralt rule 490 will modify the sufficiency and necessity functions of rule 484 to change the belief in a validated increasing sodium concentration.
Thus having means for obtaining validated sensor indications, the computer control system establishes a third type of subsystem which utilizes the validated sensor indications to come to some valid conclusion relative to components of the operating system being diagnosed. For the steam turbine power plant of FIG. 1, validated sensor readings are obtained at the influent and effluent of a particular component to make the malfunction diagnosis.
By way of example, FIG. 10 illustrates a third type of subsystem which can diagnose whether or not condenser 36 has a leak, the consequence of which would be to allow cooling water to mix with the condensate contaminating it.
Very basically, utilizing the previously described subsystems, validated anion and sodium concentrations of the influent to condenser 36 are obtained by sensors of array 50 (FIG. 1) connected to the steam line and validated anion and sodium concentrations of the effluent of condenser 36 are obtained by sensors of array 54 connected to the condensate line.
In FIG. 10, sensors 496 and 497 are respective cation conductivity and sodium sensors connected to the steam path while sensors 498 and 499 are respective cation conductivity and sodium sensors connected in the condensate path. Since the steam turbine is chemically inactive, steam and condensate sensor readings should be identical in the absence of any condenser leak. With respect to the cation conductivity reading, node 500 receives respective condensate and steam sensor values via rules 501 and 502 to compute the difference therebetween. If the results are positive by more than a predetermined amount, the belief is propagated via rule 504 to node 506 which establishes that the anion concentration in the condensate is greater than the anion concentration in the steam, a condition which may indicate a leaky condenser since the anion concentrations should be equal.
Sodium sensor readings are propagated by rules 508 and 509 to node 510 to obtain a difference in the sodium concentrations in the condensate and steam. Since the sodium sensor provides such a wide dynamic range, the relative difference is obtained at node 512 to make sure that the difference is significant. The ratio is obtained by dividing the difference propagated via rule 514 by the actual sensor reading propagated by rule 516. If the relative difference is significant, propagation to node 518 via rule 520 results in the determination that the sodium in the condensate is greater than the sodium in steam, thereby lending to the concern that there is a condenser leak since the two readings should be similar.
Other factors influencing the determination of a condenser leak would be the existence of a validated high or increasing sodium concentration in the condensate, as depicted at nodes 522 and 524 as well as a validated high or increasing anion level in the condensate as depicted at nodes 526 and 528. These latter four node results would come from identical subsystems as described in FIGS. 8 and 6 as applied to the sensors of FIGS. 10.
Another type of sensor which may be utilized in the plant of FIG. 1 provides an indication of turbine load. The subsystem of FIG. 5 as applied to the load sensor would reveal information including that illustrated at node 530 indicative of a decreasing load. If the condenser is leaking, the leak would be at a constant rate. With less load, there is less steam being condensed and the dilution of the leak is reduced such that the anion concentration in the condensate will increase. AND function 532 is responsive to a decreasing load as well as a validated increasing anion concentration to propagate, via rule 534 the belief that the anion concentration is increasing with decreasing load, as specified at node 536.
The conditions propagated via rules 538 through 544 to condenser leak malfunction node 546 individually do no more than create concern, some more than others. Conversely, the absence of such conditions influence the negative belief in a condenser leak. Collectively, however, strong evidence of the condenser leak malfunction (or lack thereof) may be obtained and presented on display 60.
Accordingly, diagnostic apparatus has been described which utilizes, in a preferred embodiment, an expert system to establish a first type of subsystem which obtains certain information relative to the output of an associated sensor. This information may then subsequently be used in a second type of subsystem which obtains an indication as to the validity of that sensor's output conclusions. A third type of subsystem utilizes validated sensor indications to obtain malfunction indications of the system being diagnosed. In the example provided, valid data relative to the influent and effluent of the condenser was utilized to diagnose a possible condenser leak. The same principles would apply to other components of the system such as the determination of condensate polisher exhaustion from polishers 40 of FIG. 1 by obtaining validated data relative to its influent and effluent by sensors of arrays 54 and 55. The establishment of these subsystems allows simultaneous determination of these component malfunctions as well as the determination of malfunctions in the sensor themselves.
Although FIG. 1 illustrates the sensor data being directly connected to the diagnostic computer 58, it is to be understood that such data at the plant could initially be collected and stored at the plant for subsequent transmission to a remote location where the diagnosis would be performed, such as described and claimed in copending application Ser. No. 605,703, filed Apr. 30, 1984, now U.S. Pat. No. 4,517,468.
FIG. 11 illustrates one of many different displays which may be utilized to convey malfunction information to a system operator. The left-hand side of display 60 lists all possible malfunctions M1 to Mn. These malfunctions would be spelled out in an actual display. Confidence in the malfunction is displayed as a horizontal bar which can occupy the scale between -1 and +1 on the display.
If the display is in color, any bar ranging from the zero position (the vertical line) in a negative direction or in a positive direction up to a distance of a (see scale at top of the figure) may be displayed as the color green indicating little or no concern. A confidence factor calculated to have a value greater than a but less than b may be displayed in a second color such as yellow, signifying a situation of some concern. Confidence factors calculated to have a value greater than b may be displayed in a third color such as red, indicative of great concern. For the display example of FIG. 11 therefor, malfunctions 1, 2, and n would be of no concern to the operator, malfunction M3 represents a condition which should be watched, and malfunction M4 a condition for which action may need to be taken. By way of example, distance a may correspond to a confidence factor of 0.3 and distance b may correspond to a confidence factor of 0.5.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4302813 *||Feb 22, 1979||Nov 24, 1981||Hitachi, Ltd.||Method of controlling operation of rotary machines by diagnosing abnormal conditions|
|US4328556 *||Apr 16, 1979||May 4, 1982||Tokyo Denryoku Kabushiki Kaisha||Control system of plants by means of electronic computers|
|US4389706 *||May 5, 1972||Jun 21, 1983||Westinghouse Electric Corp.||Digital computer monitored and/or operated system or process which is structured for operation with an improved automatic programming process and system|
|US4404637 *||Apr 30, 1981||Sep 13, 1983||Phillips Petroleum Company||Process control system|
|US4471446 *||Jul 12, 1982||Sep 11, 1984||Westinghouse Electric Corp.||Control system and method for a steam turbine having a steam bypass arrangement|
|US4517468 *||Apr 30, 1984||May 14, 1985||Westinghouse Electric Corp.||Diagnostic system and method|
|1||*||Fox et al: Techniques for Sensor Based Diagnosis, Proceedings of the Eighth International Joint Conference on Artificial Intelligence, IJCAI 83, vol. 1, pp. 158 163.|
|2||Fox et al: Techniques for Sensor Based Diagnosis, Proceedings of the Eighth International Joint Conference on Artificial Intelligence, IJCAI-83, vol. 1, pp. 158-163.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4754409 *||Jun 26, 1985||Jun 28, 1988||International Business Machines Corporation||Method for dynamically collecting current data from specified external processes and procedures for use in an expert system|
|US4754410 *||Feb 6, 1986||Jun 28, 1988||Westinghouse Electric Corp.||Automated rule based process control method with feedback and apparatus therefor|
|US4763277 *||Jan 17, 1986||Aug 9, 1988||International Business Machines Corporation||Method for obtaining information in an expert system|
|US4766550 *||Oct 30, 1985||Aug 23, 1988||Westinghouse Electric Corp.||Automatic on-line chemistry monitoring system|
|US4783752 *||Mar 6, 1986||Nov 8, 1988||Teknowledge, Inc.||Knowledge based processor for application programs using conventional data processing capabilities|
|US4792911 *||Jan 17, 1986||Dec 20, 1988||Westinghouse Electric Corp.||Diagnostic apparatus for an electric generator seal oil system|
|US4825195 *||May 1, 1987||Apr 25, 1989||Yves Berruyer||Method of monitoring to anticipate the triggering of an alarm|
|US4839822 *||Aug 13, 1987||Jun 13, 1989||501 Synthes (U.S.A.)||Computer system and method for suggesting treatments for physical trauma|
|US4839823 *||Aug 10, 1988||Jun 13, 1989||Kabushiki Kaisha Toshiba||Automatic trouble analysis apparatus and method thereof|
|US4841456 *||Sep 9, 1986||Jun 20, 1989||The Boeing Company||Test system and method using artificial intelligence control|
|US4885705 *||Feb 25, 1988||Dec 5, 1989||Eastman Kodak Company||Expert system shell for building photofinishing diagnostic systems|
|US4901229 *||Jan 21, 1986||Feb 13, 1990||Tsutomu Tashiro||Parallelized rules processing system using associative memory for pipelined execution of plural join operations and concurrent condition comparing|
|US4907167 *||Sep 30, 1987||Mar 6, 1990||E. I. Du Pont De Nemours And Company||Process control system with action logging|
|US4910691 *||Sep 30, 1987||Mar 20, 1990||E.I. Du Pont De Nemours & Co.||Process control system with multiple module sequence options|
|US4916628 *||Jul 8, 1988||Apr 10, 1990||Commonwealth Edison Company||Microprocessor-based control/status monitoring arrangement|
|US4931950 *||Jul 25, 1988||Jun 5, 1990||Electric Power Research Institute||Multimedia interface and method for computer system|
|US4951234 *||Jun 22, 1988||Aug 21, 1990||Westinghouse Electric Corp.||Monitoring a plurality of identical process parameters|
|US4967337 *||Oct 11, 1988||Oct 30, 1990||Texas Instruments Incorporated||Automated diagnostic system|
|US4975865 *||May 31, 1989||Dec 4, 1990||Mitech Corporation||Method and apparatus for real-time control|
|US4979124 *||Oct 5, 1988||Dec 18, 1990||Cornell Research Foundation||Adaptive, neural-based signal processor|
|US4985857 *||Aug 19, 1988||Jan 15, 1991||General Motors Corporation||Method and apparatus for diagnosing machines|
|US5014220 *||Sep 6, 1988||May 7, 1991||The Boeing Company||Reliability model generator|
|US5058043 *||Apr 5, 1989||Oct 15, 1991||E. I. Du Pont De Nemours & Co. (Inc.)||Batch process control using expert systems|
|US5063527 *||Mar 26, 1990||Nov 5, 1991||British Gas Plc||Monitor system|
|US5067099 *||Apr 10, 1989||Nov 19, 1991||Allied-Signal Inc.||Methods and apparatus for monitoring system performance|
|US5070468 *||Oct 12, 1990||Dec 3, 1991||Mitsubishi Jukogyo Kabushiki Kaisha||Plant fault diagnosis system|
|US5081598 *||Feb 21, 1989||Jan 14, 1992||Westinghouse Electric Corp.||Method for associating text in automatic diagnostic system to produce recommended actions automatically|
|US5089978 *||Feb 9, 1990||Feb 18, 1992||Westinghouse Electric Corp.||Automatic plant state diagnosis system including a display selection system for selecting displays responsive to the diagnosis|
|US5090014 *||Nov 1, 1989||Feb 18, 1992||Digital Equipment Corporation||Identifying likely failure points in a digital data processing system|
|US5099436 *||Nov 3, 1988||Mar 24, 1992||Allied-Signal Inc.||Methods and apparatus for performing system fault diagnosis|
|US5101337 *||Nov 7, 1989||Mar 31, 1992||Mitsubishi Denki Kabushiki Kaisha||Plant diagnostic apparatus|
|US5119318 *||Apr 17, 1989||Jun 2, 1992||Del Partners L.P.||Expert control system for real time management of automated factory equipment|
|US5121496 *||Jul 25, 1988||Jun 9, 1992||Westinghouse Electric Corp.||Method for creating, maintaining and using an expert system by recursively modifying calibration file and merging with standard file|
|US5122976 *||Mar 12, 1990||Jun 16, 1992||Westinghouse Electric Corp.||Method and apparatus for remotely controlling sensor processing algorithms to expert sensor diagnoses|
|US5130936 *||Sep 14, 1990||Jul 14, 1992||Arinc Research Corporation||Method and apparatus for diagnostic testing including a neural network for determining testing sufficiency|
|US5132920 *||Feb 16, 1988||Jul 21, 1992||Westinghouse Electric Corp.||Automated system to prioritize repair of plant equipment|
|US5193143 *||Nov 7, 1989||Mar 9, 1993||Honeywell Inc.||Problem state monitoring|
|US5228116 *||May 17, 1991||Jul 13, 1993||Aicorp., Inc.||Knowledge base management system|
|US5265035 *||May 18, 1992||Nov 23, 1993||The University Of Chicago||System diagnostics using qualitative analysis and component functional classification|
|US5274572 *||Mar 6, 1990||Dec 28, 1993||Schlumberger Technology Corporation||Method and apparatus for knowledge-based signal monitoring and analysis|
|US5293323 *||Oct 24, 1991||Mar 8, 1994||General Electric Company||Method for fault diagnosis by assessment of confidence measure|
|US5305426 *||May 14, 1992||Apr 19, 1994||Kabushiki Kaisha Toshiba||Plant operation support system for diagnosing malfunction of plant|
|US5329465 *||Aug 22, 1989||Jul 12, 1994||Westinghouse Electric Corp.||Online valve diagnostic monitoring system|
|US5345383 *||Nov 10, 1993||Sep 6, 1994||Caterpillar Inc.||Method and apparatus for selectively monitoring input|
|US5351200 *||Nov 22, 1991||Sep 27, 1994||Westinghouse Electric Corporation||Process facility monitor using fuzzy logic|
|US5406476 *||Sep 23, 1993||Apr 11, 1995||Sun Microsystems, Inc.||Method and apparatus for resource constraint scheduling|
|US5406502 *||Jun 29, 1993||Apr 11, 1995||Elbit Ltd.||System and method for measuring the operation of a device|
|US5418887 *||Apr 28, 1994||May 23, 1995||U.S. Philips Corporation||Data processing apparatus for a temporal based rule system|
|US5442555 *||Oct 7, 1993||Aug 15, 1995||Argonne National Laboratory||Combined expert system/neural networks method for process fault diagnosis|
|US5469463 *||May 8, 1991||Nov 21, 1995||Digital Equipment Corporation||Expert system for identifying likely failure points in a digital data processing system|
|US5479560 *||Oct 27, 1993||Dec 26, 1995||Technology Research Association Of Medical And Welfare Apparatus||Formant detecting device and speech processing apparatus|
|US5488697 *||Mar 8, 1994||Jan 30, 1996||Honeywell Inc.||Problem state monitoring system|
|US5490089 *||May 2, 1995||Feb 6, 1996||Xerox Corporation||Interactive user support system and method using sensors and machine knowledge|
|US5511004 *||Jun 2, 1993||Apr 23, 1996||Thomson-Csf||Diagnostic method for an evolutionary process|
|US5550751 *||Oct 15, 1993||Aug 27, 1996||The Texas A & M University System||Expert system for detecting high impedance faults|
|US5566091 *||Jun 30, 1994||Oct 15, 1996||Caterpillar Inc.||Method and apparatus for machine health inference by comparing two like loaded components|
|US5598511 *||Dec 28, 1992||Jan 28, 1997||Intel Corporation||Method and apparatus for interpreting data and accessing on-line documentation in a computer system|
|US5608845 *||Nov 29, 1993||Mar 4, 1997||Hitachi, Ltd.||Method for diagnosing a remaining lifetime, apparatus for diagnosing a remaining lifetime, method for displaying remaining lifetime data, display apparatus and expert system|
|US5629862 *||Aug 30, 1994||May 13, 1997||Electric Power Research Institute||Rule-based procedure for automatic selection of contingencies in evaluation of dynamic security of a power distribution system|
|US5696696 *||Oct 11, 1994||Dec 9, 1997||Betzdearborn, Inc.||Apparatus and method for automatically achieving and maintaining congruent control in an industrial boiler|
|US5710723 *||Apr 5, 1995||Jan 20, 1998||Dayton T. Brown||Method and apparatus for performing pre-emptive maintenance on operating equipment|
|US5748495 *||May 30, 1995||May 5, 1998||Hitachi, Ltd.||Plant operating and monitoring apparatus|
|US5777896 *||Jun 9, 1997||Jul 7, 1998||Hitachi, Ltd.||Plant operating and monitoring apparatus|
|US5835886 *||Apr 28, 1997||Nov 10, 1998||Siemens Aktiengesellschaft||Method for analyzing a measurement value and measurement value analyzer for carrying out the method|
|US5845272 *||Nov 29, 1996||Dec 1, 1998||General Electric Company||System and method for isolating failures in a locomotive|
|US5905989 *||Nov 27, 1996||May 18, 1999||Bently Nevada Corporation||Knowledge manager relying on a hierarchical default expert system: apparatus and method|
|US5923571 *||Oct 6, 1997||Jul 13, 1999||Betzdearborn, Inc.||Apparatus and method for automatic congruent control of multiple boilers sharing a common feedwater line and chemical feed point|
|US5950147 *||Jun 5, 1997||Sep 7, 1999||Caterpillar Inc.||Method and apparatus for predicting a fault condition|
|US5956663 *||Mar 26, 1998||Sep 21, 1999||Rosemount, Inc.||Signal processing technique which separates signal components in a sensor for sensor diagnostics|
|US6017143 *||Mar 28, 1996||Jan 25, 2000||Rosemount Inc.||Device in a process system for detecting events|
|US6047220 *||Dec 29, 1997||Apr 4, 2000||Rosemount Inc.||Device in a process system for validating a control signal from a field device|
|US6119047 *||Nov 10, 1997||Sep 12, 2000||Rosemount Inc.||Transmitter with software for determining when to initiate diagnostics|
|US6175934||Dec 15, 1997||Jan 16, 2001||General Electric Company||Method and apparatus for enhanced service quality through remote diagnostics|
|US6298454||Feb 22, 1999||Oct 2, 2001||Fisher-Rosemount Systems, Inc.||Diagnostics in a process control system|
|US6356191||Jun 17, 1999||Mar 12, 2002||Rosemount Inc.||Error compensation for a process fluid temperature transmitter|
|US6356857 *||Oct 27, 1998||Mar 12, 2002||Aspen Technology, Inc.||Sensor validation apparatus and method|
|US6363330||Oct 9, 1998||Mar 26, 2002||Satnam Singh Sampuran Alag||Thermocouple failure detection in power generation turbines|
|US6370448||Oct 12, 1998||Apr 9, 2002||Rosemount Inc.||Communication technique for field devices in industrial processes|
|US6397114||May 3, 1999||May 28, 2002||Rosemount Inc.||Device in a process system for detecting events|
|US6421571||Feb 29, 2000||Jul 16, 2002||Bently Nevada Corporation||Industrial plant asset management system: apparatus and method|
|US6434504||Aug 6, 1999||Aug 13, 2002||Rosemount Inc.||Resistance based process control device diagnostics|
|US6446027 *||Sep 17, 1999||Sep 3, 2002||General Electric Company||Intelligent analysis system and method for fluid-filled electrical equipment|
|US6449574||Jul 14, 2000||Sep 10, 2002||Micro Motion, Inc.||Resistance based process control device diagnostics|
|US6473710||Jun 29, 2000||Oct 29, 2002||Rosemount Inc.||Low power two-wire self validating temperature transmitter|
|US6494617||Apr 30, 1999||Dec 17, 2002||General Electric Company||Status detection apparatus and method for fluid-filled electrical equipment|
|US6505517||Jul 23, 1999||Jan 14, 2003||Rosemount Inc.||High accuracy signal processing for magnetic flowmeter|
|US6519546||Oct 19, 1998||Feb 11, 2003||Rosemount Inc.||Auto correcting temperature transmitter with resistance based sensor|
|US6532392||Jul 28, 2000||Mar 11, 2003||Rosemount Inc.||Transmitter with software for determining when to initiate diagnostics|
|US6539267||May 4, 2000||Mar 25, 2003||Rosemount Inc.||Device in a process system for determining statistical parameter|
|US6556145||Sep 24, 1999||Apr 29, 2003||Rosemount Inc.||Two-wire fluid temperature transmitter with thermocouple diagnostics|
|US6557118||Mar 8, 2001||Apr 29, 2003||Fisher Rosemount Systems Inc.||Diagnostics in a process control system|
|US6594603||Sep 30, 1999||Jul 15, 2003||Rosemount Inc.||Resistive element diagnostics for process devices|
|US6601005||Jun 25, 1999||Jul 29, 2003||Rosemount Inc.||Process device diagnostics using process variable sensor signal|
|US6611775||May 23, 2000||Aug 26, 2003||Rosemount Inc.||Electrode leakage diagnostics in a magnetic flow meter|
|US6615090||Feb 7, 2000||Sep 2, 2003||Fisher-Rosemont Systems, Inc.||Diagnostics in a process control system which uses multi-variable control techniques|
|US6615149||May 23, 2000||Sep 2, 2003||Rosemount Inc.||Spectral diagnostics in a magnetic flow meter|
|US6629059||Mar 12, 2002||Sep 30, 2003||Fisher-Rosemount Systems, Inc.||Hand held diagnostic and communication device with automatic bus detection|
|US6633782||Feb 7, 2000||Oct 14, 2003||Fisher-Rosemount Systems, Inc.||Diagnostic expert in a process control system|
|US6654697||Aug 27, 1999||Nov 25, 2003||Rosemount Inc.||Flow measurement with diagnostics|
|US6701274||Aug 27, 1999||Mar 2, 2004||Rosemount Inc.||Prediction of error magnitude in a pressure transmitter|
|US6735484||Sep 20, 2000||May 11, 2004||Fargo Electronics, Inc.||Printer with a process diagnostics system for detecting events|
|US6754601||Sep 30, 1999||Jun 22, 2004||Rosemount Inc.||Diagnostics for resistive elements of process devices|
|US6772036||Aug 30, 2001||Aug 3, 2004||Fisher-Rosemount Systems, Inc.||Control system using process model|
|US6775576||Jul 8, 2002||Aug 10, 2004||Bently Nevada, Llc||Industrial plant asset management system: apparatus and method|
|US6889096 *||Jul 15, 2002||May 3, 2005||Bently Nevada, Llc||Industrial plant asset management system: apparatus and method|
|US6898554 *||Nov 30, 2000||May 24, 2005||Scientific Monitoring, Inc.||Fault detection in a physical system|
|US6907383||May 9, 2001||Jun 14, 2005||Rosemount Inc.||Flow diagnostic system|
|US6909990 *||Feb 12, 2003||Jun 21, 2005||Kabushiki Kaisha Toshiba||Method and system for diagnosis of plant|
|US6920799||Apr 15, 2004||Jul 26, 2005||Rosemount Inc.||Magnetic flow meter with reference electrode|
|US6934696||Sep 15, 2000||Aug 23, 2005||Bently Nevada, Llc||Custom rule system and method for expert systems|
|US6970003||Mar 5, 2001||Nov 29, 2005||Rosemount Inc.||Electronics board life prediction of microprocessor-based transmitters|
|US7010459||Jun 5, 2003||Mar 7, 2006||Rosemount Inc.||Process device diagnostics using process variable sensor signal|
|US7018800||Aug 7, 2003||Mar 28, 2006||Rosemount Inc.||Process device with quiescent current diagnostics|
|US7046180||Apr 21, 2004||May 16, 2006||Rosemount Inc.||Analog-to-digital converter with range error detection|
|US7085610||Oct 5, 2001||Aug 1, 2006||Fisher-Rosemount Systems, Inc.||Root cause diagnostics|
|US7254518||Mar 15, 2004||Aug 7, 2007||Rosemount Inc.||Pressure transmitter with diagnostics|
|US7290450||Jul 16, 2004||Nov 6, 2007||Rosemount Inc.||Process diagnostics|
|US7321846||Oct 5, 2006||Jan 22, 2008||Rosemount Inc.||Two-wire process control loop diagnostics|
|US7451003||Mar 4, 2004||Nov 11, 2008||Falconeer Technologies Llc||Method and system of monitoring, sensor validation and predictive fault analysis|
|US7523667||Dec 23, 2003||Apr 28, 2009||Rosemount Inc.||Diagnostics of impulse piping in an industrial process|
|US7590511||Sep 25, 2007||Sep 15, 2009||Rosemount Inc.||Field device for digital process control loop diagnostics|
|US7623932||Dec 20, 2005||Nov 24, 2009||Fisher-Rosemount Systems, Inc.||Rule set for root cause diagnostics|
|US7627441||Sep 30, 2003||Dec 1, 2009||Rosemount Inc.||Process device with vibration based diagnostics|
|US7630861||May 25, 2006||Dec 8, 2009||Rosemount Inc.||Dedicated process diagnostic device|
|US7750642||Sep 28, 2007||Jul 6, 2010||Rosemount Inc.||Magnetic flowmeter with verification|
|US7921734||May 12, 2009||Apr 12, 2011||Rosemount Inc.||System to detect poor process ground connections|
|US7940189||Sep 26, 2006||May 10, 2011||Rosemount Inc.||Leak detector for process valve|
|US7949495||Aug 17, 2005||May 24, 2011||Rosemount, Inc.||Process variable transmitter with diagnostics|
|US7953501||Sep 25, 2006||May 31, 2011||Fisher-Rosemount Systems, Inc.||Industrial process control loop monitor|
|US8069667||Feb 6, 2009||Dec 6, 2011||Siemens Energy, Inc.||Deaerator apparatus in a superatmospheric condenser system|
|US8112565||Jun 6, 2006||Feb 7, 2012||Fisher-Rosemount Systems, Inc.||Multi-protocol field device interface with automatic bus detection|
|US8112997||Feb 6, 2009||Feb 14, 2012||Siemens Energy, Inc.||Condensate polisher circuit|
|US8290721||Aug 14, 2006||Oct 16, 2012||Rosemount Inc.||Flow measurement diagnostics|
|US8417432||Apr 30, 2008||Apr 9, 2013||United Technologies Corporation||Method for calculating confidence on prediction in fault diagnosis systems|
|US8788070||Sep 26, 2006||Jul 22, 2014||Rosemount Inc.||Automatic field device service adviser|
|US8886582||Feb 17, 2011||Nov 11, 2014||Snecma||Failure detection and locating system and engine containing the same|
|US8898036||Aug 6, 2007||Nov 25, 2014||Rosemount Inc.||Process variable transmitter with acceleration sensor|
|US8964338||Jan 9, 2013||Feb 24, 2015||Emerson Climate Technologies, Inc.||System and method for compressor motor protection|
|US8974573||Mar 15, 2013||Mar 10, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9017461||Mar 15, 2013||Apr 28, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9021819||Mar 15, 2013||May 5, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9023136||Mar 15, 2013||May 5, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9046900||Feb 14, 2013||Jun 2, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring refrigeration-cycle systems|
|US9052240||Jun 29, 2012||Jun 9, 2015||Rosemount Inc.||Industrial process temperature transmitter with sensor stress diagnostics|
|US9081394||Mar 15, 2013||Jul 14, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9086704||Mar 15, 2013||Jul 21, 2015||Emerson Climate Technologies, Inc.||Method and apparatus for monitoring a refrigeration-cycle system|
|US9121407||Jul 1, 2013||Sep 1, 2015||Emerson Climate Technologies, Inc.||Compressor diagnostic and protection system and method|
|US9140728||Oct 30, 2008||Sep 22, 2015||Emerson Climate Technologies, Inc.||Compressor sensor module|
|US9163828||Oct 31, 2011||Oct 20, 2015||Emerson Process Management Power & Water Solutions, Inc.||Model-based load demand control|
|US9194894||Feb 19, 2013||Nov 24, 2015||Emerson Climate Technologies, Inc.||Compressor sensor module|
|US9207129||Sep 27, 2012||Dec 8, 2015||Rosemount Inc.||Process variable transmitter with EMF detection and correction|
|US9207670||Sep 19, 2011||Dec 8, 2015||Rosemount Inc.||Degrading sensor detection implemented within a transmitter|
|US9217565 *||Feb 7, 2011||Dec 22, 2015||Emerson Process Management Power & Water Solutions, Inc.||Dynamic matrix control of steam temperature with prevention of saturated steam entry into superheater|
|US9285802||Feb 28, 2012||Mar 15, 2016||Emerson Electric Co.||Residential solutions HVAC monitoring and diagnosis|
|US9304521||Oct 7, 2011||Apr 5, 2016||Emerson Climate Technologies, Inc.||Air filter monitoring system|
|US9310094||Feb 8, 2012||Apr 12, 2016||Emerson Climate Technologies, Inc.||Portable method and apparatus for monitoring refrigerant-cycle systems|
|US9310439||Sep 23, 2013||Apr 12, 2016||Emerson Climate Technologies, Inc.||Compressor having a control and diagnostic module|
|US9335042||Aug 16, 2010||May 10, 2016||Emerson Process Management Power & Water Solutions, Inc.||Steam temperature control using dynamic matrix control|
|US9348722 *||Oct 21, 2013||May 24, 2016||Round Rock Research, Llc||Diagnostic and managing distributed processor system|
|US9447963||Feb 7, 2011||Sep 20, 2016||Emerson Process Management Power & Water Solutions, Inc.||Dynamic tuning of dynamic matrix control of steam temperature|
|US20020066054 *||Nov 30, 2000||May 30, 2002||Jaw Link C.||Fault detection in a physical system|
|US20030028269 *||Jul 15, 2002||Feb 6, 2003||Bob Spriggs||Industrial plant asset management system: apparatus and method|
|US20030154051 *||Feb 12, 2003||Aug 14, 2003||Kabushiki Kaisha Toshiba||Method and system for diagnosis of plant|
|US20040024568 *||Jun 5, 2003||Feb 5, 2004||Evren Eryurek||Process device diagnostics using process variable sensor signal|
|US20040249583 *||Mar 15, 2004||Dec 9, 2004||Evren Eryurek||Pressure transmitter with diagnostics|
|US20050011278 *||Jul 16, 2004||Jan 20, 2005||Brown Gregory C.||Process diagnostics|
|US20050030185 *||Aug 7, 2003||Feb 10, 2005||Huisenga Garrie D.||Process device with quiescent current diagnostics|
|US20050072239 *||Sep 30, 2003||Apr 7, 2005||Longsdorf Randy J.||Process device with vibration based diagnostics|
|US20050132808 *||Dec 23, 2003||Jun 23, 2005||Brown Gregory C.||Diagnostics of impulse piping in an industrial process|
|US20050210337 *||Mar 4, 2004||Sep 22, 2005||Falconeer Technologies Llc||Method and system of monitoring, sensor validation and predictive fault analysis|
|US20060036404 *||Aug 17, 2005||Feb 16, 2006||Wiklund David E||Process variable transmitter with diagnostics|
|US20060095394 *||Dec 20, 2005||May 4, 2006||Miller John P||Rule set for root cause diagnostics|
|US20060277000 *||Aug 14, 2006||Dec 7, 2006||Wehrs David L||Flow measurement diagnostics|
|US20060282580 *||Jun 6, 2006||Dec 14, 2006||Russell Alden C Iii||Multi-protocol field device interface with automatic bus detection|
|US20070010968 *||May 25, 2006||Jan 11, 2007||Longsdorf Randy J||Dedicated process diagnostic device|
|US20070068225 *||Sep 29, 2005||Mar 29, 2007||Brown Gregory C||Leak detector for process valve|
|US20070150305 *||Feb 1, 2005||Jun 28, 2007||Klaus Abraham-Fuchs||Method for selecting a potential participant for a medical study on the basis of a selection criterion|
|US20080125884 *||Sep 26, 2006||May 29, 2008||Schumacher Mark S||Automatic field device service adviser|
|US20090043405 *||Oct 3, 2008||Feb 12, 2009||Chester Daniel L||Method and system of monitoring, sensor validation and predictive fault analysis|
|US20090043530 *||Aug 6, 2007||Feb 12, 2009||Sittler Fred C||Process variable transmitter with acceleration sensor|
|US20090083001 *||Sep 25, 2007||Mar 26, 2009||Huisenga Garrie D||Field device for digital process control loop diagnostics|
|US20090266076 *||Feb 6, 2009||Oct 29, 2009||Siemens Energy, Inc.||Condensate Polisher Circuit|
|US20090276136 *||Apr 30, 2008||Nov 5, 2009||Steven Wayne Butler||Method for calculating confidence on prediction in fault diagnosis systems|
|US20090303057 *||Sep 26, 2006||Dec 10, 2009||Brown Gregory C||Leak detector for process valve|
|US20100017092 *||Jul 16, 2008||Jan 21, 2010||Steven Wayne Butler||Hybrid fault isolation system utilizing both model-based and empirical components|
|US20100199671 *||Feb 6, 2009||Aug 12, 2010||Siemens Energy, Inc.||Deaerator Apparatus in a Superatmospheric Condenser System|
|US20100288054 *||May 12, 2009||Nov 18, 2010||Foss Scot R||System to detect poor process ground connections|
|US20120040299 *||Feb 7, 2011||Feb 16, 2012||Emerson Process Management Power & Water Solutions, Inc.||Dynamic matrix control of steam temperature with prevention of saturated steam entry into superheater|
|US20140115404 *||Oct 21, 2013||Apr 24, 2014||Round Rock Research Llc||Diagnostic and managing distributed processor system|
|EP0347645A2 *||Jun 7, 1989||Dec 27, 1989||Westinghouse Electric Corporation||Monitoring a plurality of parameters from identical processes|
|WO2011104466A1 *||Feb 17, 2011||Sep 1, 2011||Snecma||System for detecting and locating engine breakdowns and engine comprising such a system|
|U.S. Classification||702/185, 706/915, 706/52, 700/79|
|International Classification||F01K7/24, G06F9/06, G06F11/30, F22B35/18, F02G1/00, G06F19/00, G06F17/00, H04B3/00, G05B23/02, F01K13/00|
|Cooperative Classification||Y10S706/915, G05B23/0278|
|Nov 5, 1984||AS||Assignment|
Owner name: WESTINGHOUSE ELECTRIC CORPORATION WESTINGHOUSE BLD
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:KEMPER, CHRISTIAN T.;BELLOWS, JAMES C.;KLEINOSKY, PAMELA J.;REEL/FRAME:004333/0150
Effective date: 19840724
|Oct 27, 1987||CC||Certificate of correction|
|Nov 2, 1988||AS||Assignment|
Owner name: HUGHES TOOL COMPANY
Free format text: CHANGE OF NAME;ASSIGNOR:HUGHES TOOL COMPANY-USA, A CORP. OF DE;REEL/FRAME:005169/0319
Effective date: 19881006
|Mar 5, 1990||FPAY||Fee payment|
Year of fee payment: 4
|Jun 9, 1994||FPAY||Fee payment|
Year of fee payment: 8
|Sep 8, 1998||REMI||Maintenance fee reminder mailed|
|Feb 14, 1999||LAPS||Lapse for failure to pay maintenance fees|
|Apr 27, 1999||FP||Expired due to failure to pay maintenance fee|
Effective date: 19990217